KR20220051136A - Method, apparatus, apparatus and computer recording medium for generating broadcast voice - Google Patents

Abstract

The present application discloses a method, an apparatus, an apparatus, and a computer recording medium for generating a broadcast voice in the fields of voice technology and knowledge graph technology. A specific implementation method is to obtain a speech matching a scene from a voice package, obtain a broadcast template pre-arranged for the scene, and use the speech to fill the broadcast template to generate a broadcast voice. According to the present application, the broadcast voice can realize the personality characteristics of the entity target of the voice package well, and the broadcast effect can be further improved.

Description

Method, apparatus, apparatus and computer recording medium for generating broadcast voice

The present application relates to the field of computer application technology, and more particularly, to a method, an apparatus, an apparatus, and a computer recording medium for generating a broadcast voice in a voice technology and a knowledge graph technology.

<Cross-reference to related applications>

This application has an application date of October 15, 2020, an application number of 2020111059358, and the title of the invention is "method and device for generating broadcast voice, electronic equipment, and computer storage medium)", claiming the priority of Chinese patent applications.

As the user's demand for the function of the smart terminal increases, the function of voice broadcasting is integrated into more and more application programs. The user can download and install various voice packages so that the voice broadcast can use the sound of his or her favorite task.

Currently, voice broadcasting satisfies users' demands in terms of sound to a considerable degree, but the effect is not satisfactory because the contents of voice broadcasting are all fixed in each scene. For example, when navigation is started, no matter which voice package the user uses, all broadcast "start of departure".

In consideration of this, the present application may provide a method, an apparatus, an apparatus, and a computer recording medium for generating a broadcast voice for facilitating improving the effect of voice broadcasting.

In a first aspect, the present application provides a method for generating a broadcast voice,

obtaining a narration matching a scene from a voice package, and obtaining a broadcast template pre-configured for the scene; and

and generating a broadcast voice by charging the broadcast template using the speech.

In a second aspect, the present application provides an apparatus for generating a broadcast voice,

a narration acquisition module for acquiring a narration matching the scene from the voice package;

a template obtaining module for obtaining a broadcast template preconfigured for the scene; and

and a voice generating module configured to generate a broadcasting voice by charging the broadcasting template using the narration.

In a third aspect, the present application provides an electronic device,

at least one processor; and

a memory communicatively coupled to the at least one processor;

An instruction executable by the at least one processor is stored in the memory, and the instruction is executed by the at least one processor, so that the method is performed by the at least one processor.

In a fourth aspect, the present application provides a non-transitory computer-readable recording medium having a computer instruction stored thereon, the computer instruction causing the computer to perform the method.

It can be seen from the above technical solution that the present application uses a speech matching a scene in a voice package to obtain a broadcasting voice after filling a broadcasting template, and the broadcasting voice reflects the personality characteristics of the entity object of the voice package. It can be implemented well, further improve the broadcasting effect, and make the user really have the feeling that the entity object of the voice package is speaking.

Other effects of the selectable method will be further described in conjunction with the following specific examples.

The accompanying drawings are for better understanding of the present solution, and are not intended to limit the present application.
1 is a schematic diagram of a principle of generating a broadcast voice in the prior art.
2 shows an exemplary system architecture that may be applied to an embodiment of the present application.
3 is a flowchart of a main method provided in an embodiment of the present application.
4 is a schematic diagram of a principle of generating a broadcast voice provided in an embodiment of the present application.
5 is a flowchart of a method for discovering a style speech provided in an embodiment of the present application.
6 is a flowchart of a method for discovering a knowledge discourse provided in an embodiment of the present application.
7 is an exemplary diagram of some knowledge graphs provided in embodiments of the present application.
8 is a structural diagram of an apparatus for generating broadcast voice provided in an embodiment of the present application.
9 is a block diagram of an electronic device implementing an embodiment of the present application.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present application will be described in conjunction with the accompanying drawings, which include various details of the embodiments of the present application for easy understanding, which should be regarded as exemplary only. Accordingly, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Likewise, for clarity and brevity, descriptions of well-known functions and structures are omitted from the description below.

In the prior art, a principle of generating a broadcast voice may be the same as in FIG. 1 . The step of generating the broadcast text may include, but is not limited to, the following two situations.

One is the generation of broadcast texts based on dialogue. That is, after receiving the user's voice command, the response text generated in response to the user's voice command is used as the broadcast text. For example, receiving the user's voice command "Query coffee shop", the generated response text is "We will find the nearest cafe to you. Located on the C floor of Beijing International Building, Zhongguancun South Street, 2.1 Kilometers away." In this situation, mainly, the response text is generated after performing an analysis of the scene and user's intention based on the understanding of the dialogue.

The other actively generates the broadcast text. That is, during the audio broadcast of a specific function, the audio broadcast is actively performed. For example, during the process of navigation, broadcast texts such as "start of departure" and "turn left forward" are actively broadcast. In this situation, the broadcast text is generated mainly after scene analysis based on the current actual situation.

After generating the broadcast text, the voice to be broadcast is obtained by performing voice synthesis using the tone information in the voice package. The broadcast audio generated by the prior art is the same as the audio content broadcast by other audio packages in the same scene, only with a difference in tone. For example, the user uses the son's voice package, or the star's voice package to "look up the cafe." In the scene where they all say, "We'll find the nearest cafe. It's located at ***." to broadcast

2 shows an exemplary system architecture of a method for generating a broadcast voice or an apparatus for generating a broadcast voice that can be applied to an embodiment of the present application.

As shown in FIG. 2 , the system architecture may include terminal devices 101 and 102 , a network 103 , and a server 104 . The network 103 provides a medium for a communication link between the terminal devices 101 , 102 and the server 104 . Network 103 may include various types of connections, such as wired, wireless communication links, or fiber optic cables.

A user may interact with the server 104 via the network 103 using the terminal devices 101 , 102 . Various applications, such as a voice interaction application, a map application, a web browser application, and a communication type application, may be installed in the terminal devices 101 and 102 .

The terminal devices 101 and 102 may be various electronic devices that support voice broadcasting. Smartphones, tablets, laptops, smart wearable devices, and the like are included, but are not limited thereto. The apparatus for generating broadcast voice provided by the present application may be installed and performed in the aforementioned server 104 or may be installed and performed in the terminal devices 101 and 102 . It may be implemented in a plurality of software or software modules (eg, to provide a distributed service), or may be implemented in a single software or software module, but is not specifically limited here.

The server 104 may be a single server, or may be a server group constituted by a plurality of servers. The number of terminal devices, networks and servers in FIG. 2 is merely exemplary. According to implementation requirements, it may be any number of terminal devices, networks and servers.

3 is a flowchart of a main method provided in an embodiment of the present application. 3 , the method may include the following steps.

At 301 , a narration matching the scene is obtained from the voice package.

In the embodiment of the present application, the voice package further includes various kinds of speech information in addition to including tone information. A "discourse" can be understood in a way of speaking, and can use different modes of expression when expressing the same meaning, i.e., different speech styles. In embodiments of the present application, for the same scene, different voice packages may use different narrations. Among them, the narration includes at least one of a name narration, a style narration, and a knowledge narration. Calling speech is a way of expressing titles to users. Stylistic speech is a way of expression that uses a particular style. Knowledge discourse is an expression method based on specific knowledge content.

)", "당신(

)" 등의 기초적인 화술로 총칭할 수 있다. As an example, when the user uses the son's voice package, the title speech may use "Father". If the user uses the wife's voice package, the title speech can use "husband". Of course, there may be no title speech information in one voice package. For example, for the star's voice package, without using a title speech, all "you (

)", "you(

)", etc., can be collectively referred to as basic dialogue techniques.

"를 사용할 수 있다. 사용자가 코미디언의 음성 패키지를 사용할 경우, 구현되는 것은, 재미있는 스타일이며, 스타일 화술 "우리들은 보통 운전자입니다. F1을 운전하는 척하지 말아 주십시오. 감속하고 천천히 가 주세요.

"를 사용할 수 있다.Taking one style speech as an example, for the same scene "Speeding", when the user uses the person's voice package in the family, what is implemented is a heart-warming style, the style speech "Speeding, safe while driving" Please be careful.

" can be used. When users use the comedian's voice package, what's implemented is, it's a fun style, and a stylistic "We're usually drivers. Don't pretend to drive the F1. Please slow down and go slowly.

" can be used.

Taking one knowledge discourse as an example, for the scene "cafe", if the user uses the voice package of star A, the knowledge discourse "Please give me a cup of coffee xxx" can be used, where "xxx" is a star It may be a coffee brand advertised by A. If the user uses Star B's voice package, "xxx" in the knowledge narrative may be the coffee brand Star B advertises.

Methods of generating the various narrations in the voice package will be described in detail in the following examples.

At 302, a preconfigured broadcast template for the scene is obtained.

In an embodiment of the present application, a broadcast template may be configured in each scene in advance. The broadcast template may include at least one narration combination.

In 303, the broadcast template is filled using the obtained speech to generate a broadcast voice.

The broadcast template corresponds to the scene, the voice package has a personified narration matching the scene, and the broadcast text obtained after using the narration to fill the broadcast template is an entity object of the voice package (eg, son, wife, a certain celebrity, etc.) can be well implemented, and the broadcasting effect can be further improved, and the user can really feel that the entity object of the voice package is speaking.

After acquiring the broadcast text, further, by using the tone information in the audio package, voice synthesis may be performed to finally generate a broadcast voice, and this part is the same as in the prior art, and will not be described in detail.

As shown in FIG. 4 , in the present application, the narration information in the voice package is used in the process of generating the broadcast text. The following describes in detail a method of generating speech information in a voice package by combining embodiments.

) 등의 일반적인 호칭의 옵션을 포함한다. 사용자가 스스로 입력하도록 입력 박스를 제공할 수도 있다. With respect to the name utterance in the voice package, the user can install and obtain it. As a preferred embodiment, a component such as an input box or an option for a nickname speech may be provided to the user in the installation interface for the voice package so that the user can input or select the nickname speech. For example, for a user who uses the son's voice package, an installation interface for the voice package may be provided, and the installation interface includes "father", "mother", "grandfather", "husband", "wife", "grandmother", "maternal grandmother", "maternal grandfather",

), including general designation options. An input box may be provided for the user to input by himself/herself.

With respect to the style speech in the voice package, it is possible to obtain preset contents such as those preset by a research and development person, a service provider, and the like. However, as a preferred implementation method, the style narration can be obtained by excavating in advance through a search engine. For example, the steps shown in FIG. 5 may be used.

In 501, a search keyword is obtained by splicing using a preset style keyword and a keyword of the scene.

Style keywords can also be obtained by installing them. For example, in an installation interface for a voice package, an input box of a style keyword or an optional component may be provided to the user to allow the user to input or select. For example, an option of style keywords such as "intimate", "comic", "popular", "TikTok style" and the like may be provided in the installation interface for the voice package for the user to select.

At 502 , a style speech candidate is selected from the search result text corresponding to the search keyword.

Assuming that the current scene asks for a cafe, the scene keywords are "cafe", "coffee", and the style keyword of the voice package that the user is currently using is "warming heart", "cafe heart warms". , "Coffee warms the heart." It is possible to construct keywords, and after each search, it is possible to obtain search result texts such as titles and outlines of search results. After sorting based on the relevance of the search keywords, the search result texts arranged in the top N are obtained as style speech candidates. Among them, N is a preset positive integer.

At 503 , after modifying the style speech candidate item, the style speech is acquired.

In the present embodiment, in the modification of the style speech candidate item, the final style speech may be obtained after the research and development person adjusts, combines, and selects the style speech candidate item. You can also add a title slot to a style narrative. In addition to artificial insemination, other fertilization methods may be used.

For example, nominees for style speech: "Coffee can energize me, and if you want to sleep well, quit first.", "A sip of coffee is a bit bitter, but the sweetness afterward makes you forget the bitterness.", " Life is like a cup of coffee, bitter, sweet, sweet and enjoyable", after artificial insemination and style speech "Coffee can clear your head, but it can affect your sleep, so [title] is for relaxation. You have to be careful."

For the knowledge discourse in the voice package, preset contents may be acquired, and may be preset by, for example, a research and development person in charge, a service provider, and the like. However, as a preferred implementation method, knowledge discourse can be obtained by excavating in advance based on the knowledge graph. For example, the steps shown in FIG. 6 can be used.

At 601 , a knowledge graph related to the speech package is obtained.

In general, a voice package corresponds to a specific entity object, and is implemented with a tone color of the entity object. For example, when a user uses a voice package of a relative, the entity corresponding to the voice package is the relative. Then, for example, if the user uses the voice package of star A, the entity corresponding to the voice package is star A. Since each entity has a knowledge graph corresponding to it, in this step, the knowledge graph of the entity corresponding to the voice package may be obtained.

At 602 , a knowledge node matching the scene is obtained from the knowledge graph.

In the knowledge-based graph, each knowledge node includes both specific content and other related relationships between the knowledge nodes. Some contents of the knowledge graph shown in FIG. 7 are taken as an example. Taking the audio package of "Star A" as an example, the corresponding entity is "Star A", and in the knowledge graph, for example, "Whistleblower", "Lucing Coffee", "Central Theater Academy" It may include knowledge nodes such as ", "Hangzhou City", etc., among which the related relationship between "whistleblower" and "Star A" is "hit movie", "Luixing Coffee" and " The relationship between "Star A" is "Advertising Model", the relationship between "Central Theater Academy" and "Star A" is "Graduation University", and the relationship between "Hangzhou City" and "Star A" is "Hometown" . When a knowledge node matching a scene is obtained, the scene keyword, contents of the knowledge node, and related relationships may be matched.

At 603 , a knowledge narration corresponding to the scene is generated by using the obtained knowledge node and the narration template corresponding to the scene.

For each scene, a narration template of knowledge narration can be preset. For example, for the scene "Search in the cinema", the narrative template "Come to the cinema and see the new movie [film name] that I have performed" may be installed, and in step 602 , the knowledge node "Whistleman (吹哨人) )", fill it in the slot [movie name] in the narrative template, and create the knowledge narrative "Come to the cinema and watch the movie I just performed, Whistleblower."

In the voice package, some or all of a title speech, a style speech, and a knowledge speech may be included. As a preferred implementation method, in the step 301, when “obtain a narration matching the scene from the voice package.” First, a keyword of the scene is determined, and then a narration matching the keyword of the scene is obtained from the voice package. can be obtained Among them, matching may be based on a method of text similarity. For example, if the text similarity between the speech and the scene keyword is greater than or equal to a preset similarity threshold, matching may be considered. In this way, it is relatively close to the scene, and a more comprehensive narrative can be found.

In addition to the preferred implementation methods described above, other methods may be used. For example, the matching relationship between the dialogue and each scene may be set in advance.

The following is a combination of embodiments, and in step 302, "acquire a preconfigured broadcast template for the scene." and the implementation method of step 303 "filling the broadcast template using the acquired speech to generate a broadcast voice."

At least one broadcast template and attribute information of each broadcast template may be preconfigured for each scene. Among them, the broadcast template may include at least one narration combination, and may further include basic narration in addition to the above-described nominal narration, style narration, and knowledge narration, and the basic narration may be stored on the server side. The attribute information may include at least one of a constraint rule between priority and narration.

As an example, for the theme "Search the cafe.", six broadcast templates are installed, and the priority and constraint rules are shown in Table 1.

Assume that the user uses the son's voice package, and in the scene of "Search the cafe.", a narration matching the scene in the voice package is obtained, as follows.

Nickname Speech: Father,

Style speech: Drinking coffee can clear your head, but it can affect your sleep, so [Title] should be careful with rest.

In the broadcast templates shown in Table 1, the ones with the highest priority are selected from the ones with the lowest priority. Since there is no knowledge narrative that matches the scene, the first two templates cannot be used. In the constraint rule of the third template, since the constraint style speech must not have a title and cannot be used, the fourth template "[basic speech] [style speech]" can be used.

On the server side, you will get the basic narration of the scene "We will find the nearest cafe. It is located at ***." from the audio package, and the style narration of the scene from the audio package "Coffee can clear your head, but it will not affect your sleep." You can give it, so [Title] needs to be careful about rest.", recharge the 4th template, and finally broadcast text "I'll find the nearest cafe. It is located at ***, if you drink coffee, You can clear it, but it can affect your sleep, so fathers should be careful with rest."

After acquiring the broadcast text, the broadcast voice may be acquired by performing voice synthesis based on the tone information in the voice package. This method of generating a broadcast voice makes the user feel very warm and has a very strong personalizing effect, as if the user's voice was spoken by his own son.

The above is a detailed description of the method provided by the present application, and the following describes in detail the apparatus provided by the present application.

8 is a structural diagram of an apparatus for generating broadcast voice provided in an embodiment of the present application. The device may be located in the application of the local terminal, or it may be located in a functional unit such as a plug-in or a software development kit (SDK) in the application of the local terminal, or located on the server side may be As shown in FIG. 8 , the apparatus may include a speech obtaining module 00 , a template obtaining module 10 and a voice generating module 20 , and a first mining module 30 and a second It may further include an excavation module 40 . Among them, the main functions of each component unit are as follows.

The narration acquisition module 00 is used to acquire a narration matching the scene from the voice package.

As a preferred embodiment, the narration acquisition module 00 may determine a keyword of the scene, and acquire a narration matching the keyword of the scene from the voice package.

Among them, the narration includes at least one of a name narration, a style narration, and a knowledge narration.

The template obtaining module 10 is used to obtain a preconfigured broadcast template for a scene.

As a preferred embodiment, the template obtaining module 10 determines at least one broadcast template preconfigured for a scene and property information of each broadcast template, the broadcast template includes at least one speech combination, and the properties of each broadcast template used to select a broadcast template configured in one said scene from at least one broadcast template according to the information and audio package.

The voice generating module 20 is used to generate a broadcasting voice by filling a broadcasting template using a narration.

Specifically, the speech generation module 20 may include a text generation submodule 21 and a speech synthesis submodule 22 .

The text generation sub-module 21 is used to generate a broadcast text by filling a broadcast template using a narration.

The speech synthesis sub-module 22 is used to obtain a broadcast speech by performing speech synthesis on the broadcast text by using the tone information in the speech package.

With respect to the name utterance in the voice package, the user can install and obtain it. As a preferred embodiment, a component such as an input box or an option for a nominal speech may be provided to the user in the installation interface for the voice package so that the user may input or select the nominal speech.

For the style speech in the voice package, preset content may be acquired, and may be preset, for example, by a research and development person in charge, a service provider, and the like. However, as a preferred implementation method, the style narration may be obtained by excavating in advance with the first excavation module 30 through a search engine.

The first excavation module 30 acquires the style speech in the excavation voice package in advance by using the following method, the method comprising:

Splicing using preset style keywords and keywords in the scene to obtain search keywords,

selecting a style speech candidate from the search result text corresponding to the search keyword;

Style speech is acquired by obtaining the result of modifying the style speech candidate items. Among them, as an implementation method, the candidate items for style speech may be artificially modified.

The second discovery module 40 discovers and acquires the knowledge narration in the voice package in advance using the following method, wherein the method is:

Acquire a knowledge graph related to the voice package,

obtain a knowledge node matching the scene from the knowledge graph,

Using the acquired knowledge node and the speech template corresponding to the scene, a knowledge speech corresponding to the scene is generated.

According to an embodiment of the present application, the present application further provides an electronic device and a readable recording medium.

As shown in FIG. 9 , it is a block diagram of an electronic device implementing a method for generating a broadcast voice according to an embodiment of the present application. Electronic device means various types of digital computers such as laptop computers, desktop computers, operating platforms, personal digital assistants, servers, blade servers, large computers, and other suitable computers. Electronic devices may refer to various types of mobile devices such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The elements shown in the text, their connections and relationships, and their functions are illustrative only and do not limit the implementation of the invention described and/or claimed in this text.

As shown in FIG. 9 , the electronic device includes one or a plurality of processors 901 , a memory 902 , and an interface for connecting each member including a high-speed interface and a low-speed interface. Each member may be connected to each other using a different bus, and may be mounted on a common main board or may be mounted in other ways as needed. The processor may process instructions stored in the memory or executed within the electronic device including instructions for displaying graphical information of the GUI on an external input/output device (eg, a display device coupled to the interface) in the memory. there is. In other embodiments, multiple processors and/or multiple buses may be used with multiple memories as needed. Likewise, multiple electronic devices may be connected, each providing some necessary operation (eg, a server array, a group of blade servers, or a multiprocessor system). In FIG. 9 , one processor 901 is exemplified.

Memory 902 is a non-transitory computer-readable storage medium provided herein. Here, instructions executable by at least one processor are stored in the memory, so that the at least one processor performs the method of generating a broadcast voice provided in the present application. The non-transitory computer-readable storage medium of the present application stores computer instructions, the computer instructions causing the computer to perform the method of generating a broadcast voice provided in the present application.

The memory 902 is a non-transitory computer-readable storage medium, which stores non-transitory software programs, non-transitory computer-executable programs and modules, such as program instructions/modules, corresponding to the method for generating broadcast voices in the embodiments of the present application. can be used to store The processor 901 executes non-transitory software programs, commands and modules stored in the memory 902, thereby performing various functional applications and data processing of the server, that is, generating broadcast voice in the embodiment of the method. implement the method

The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program necessary for at least one function; The data storage area may store data constructed according to the use of an electronic device according to a method of generating a broadcast voice. In addition, the memory 902 may include a high-speed random access memory, and may further include a non-transitory memory such as at least one magnetic disk storage device, a flash device, or other non-transitory solid state storage device. In some embodiments, memory 902 may optionally include memory installed remotely with respect to processor 901 , which remote memory may be coupled to an electronic device implementing a method of generating broadcast voice over a network. there is. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile networks, and combinations thereof.

The electronic device may further include an input device 903 and an output device 904 . The processor 901 , the memory 902 , the input device 903 , and the output device 904 may be connected via a bus or other method, and in FIG. 9 , the connection via a bus is exemplified.

The input device 903 may receive input number or character information, and may generate a key signal input related to user setting and function control of an electronic device implementing a method of generating a broadcast voice, for example, touch It may be an input device such as a screen, keypad, mouse, trackpad, touchpad, pointing stick, one or more mouse buttons, trackball, joystick, or the like. The output device 904 may include a display device, an auxiliary lighting device (eg, an LED), and a tactile feedback device (eg, a vibration motor). The display device may include, but is not limited to, a liquid crystal display device (LCD), a light emitting diode (LED) display device, and a plasma display device. In some embodiments, the display device may be a touch screen.

Various embodiments of the systems and techniques described herein may be implemented in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (dedicated integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include implementation in one or more computer programs, wherein the one or more computer programs may be executed and/or interpreted in a programmable system comprising at least one programmable processor, The programmable processor may be a dedicated or general purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one It can transmit to one input device and the at least one output device.

Such computing programs (also referred to as programs, software, software applications, or code) contain machine instructions of a programmable processor, and include programming languages for high-level processes and/or objects, and/or assembly/machine languages such as Computing programs can be executed. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or apparatus (eg, for example, a magnetic disk, an optical disk, a memory, a programmable logic device (PLD)), and includes a machine-readable medium for receiving machine instructions, which are machine-readable signals. The term “machine-readable signal” means any signal for providing machine instructions and/or data to a programmable processor.

To provide for interaction with a user, a computer may implement the systems and techniques described herein, the computer comprising a display device (eg, a cathode ray tube (CRT) or liquid crystal display (LCD) for displaying information to the user) device) monitor); and a keyboard and a pointing device (eg, a mouse or track ball), wherein a user provides input to the computer through the keyboard and the pointing device. Other types of devices may also provide for interaction with a user, for example, the feedback provided to the user may be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); An input from the user may be received in any form (sound input, voice input, or tactile input).

The systems and techniques described herein include a computing system that includes a back-end member (eg, used as a data server), or a computing system that includes a middleware member (eg, an application server), or a computing system that includes a front-end member. a system (eg, a user computer having a graphical user interface or network browser through which a user may interact with embodiments of the systems and techniques described herein), or such It may be implemented in a computing system that includes any combination of back-end members, middleware members, or front-end members. Any form or medium of digital data communication (eg, a communication network) may connect the members of the system to one another. Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

A computer system may include a client and a server. A client and server are typically remote from each other and interact with each other, usually via a communication network. Creating a relationship between a client and a server through a computer program running on a corresponding computer and having a client-server relationship with each other.

It should be understood that steps may be rearranged, added, or deleted using the various forms of the process described above. For example, each step described in the present application may be performed simultaneously, sequentially, or in a different order, and as long as the technical solution disclosed in the present application can realize the desired result, the text is not limited herein.

The above specific embodiments do not limit the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims (16)

In a method for generating broadcast voice,
obtaining a narration matching a scene from a voice package, and obtaining a broadcast template pre-configured for the scene; and
generating a broadcast voice by charging the broadcast template using the speech;
containing,
How to create a broadcast voice. According to claim 1,
wherein the narration comprises at least one of a title narration, a style narration and a knowledge narration,
How to create a broadcast voice. According to claim 1,
The step of obtaining a narration matching the scene from the voice package comprises:
determining a keyword of the scene; and
Including; obtaining a speech matching the keyword of the scene from the voice package;
How to create a broadcast voice. According to claim 1,
Obtaining a broadcast template preconfigured for the scene comprises:
determining at least one broadcast template preconfigured for the scene and attribute information of each broadcast template, the broadcast template including at least one speech combination; and
Selecting a broadcast template configured for one scene from the at least one broadcast template according to attribute information of each broadcast template and the audio package;
How to create a broadcast voice. According to claim 1,
The step of generating a broadcast voice by charging the broadcast template using the narration comprises:
filling the broadcast template using the narration to generate broadcast text; and
obtaining the broadcast voice by performing voice synthesis on the broadcast text using the tone information in the voice package;
How to create a broadcast voice. 3. The method of claim 2,
The stylized speech within the audio package is:
Splicing using preset style keywords and keywords in the scene to obtain search keywords,
selecting a style speech candidate item from the search result text corresponding to the search keyword;
obtaining a style speech by obtaining a result of modifying the style speech candidate item;
obtained by excavating in advance using the method,
How to create a broadcast voice. 3. The method of claim 2,
The knowledge discourse in the voice package is:
acquiring a knowledge graph related to the voice package,
obtaining a knowledge node matching the scene from the knowledge graph;
generating a knowledge narration corresponding to the scene by using the acquired knowledge node and the narration template corresponding to the scene;
obtained by excavating in advance using the method,
How to create a broadcast voice. In a device for generating broadcast voice,
a narration acquisition module for acquiring a narration matching the scene from the voice package;
a template obtaining module for obtaining a broadcast template preconfigured for the scene; and
a voice generating module configured to generate a broadcasting voice by charging the broadcasting template using the narration;
containing,
Broadcast voice generator. 9. The method of claim 8,
wherein the narration comprises at least one of a title narration, a style narration and a knowledge narration,
Broadcast voice generator. 9. The method of claim 8,
The dialogue acquisition module is
used to determine a keyword of the scene, and to obtain a narration matching the keyword of the scene from the voice package,
Broadcast voice generator. 9. The method of claim 8,
The template acquisition module,
Determine at least one broadcast template preconfigured for the scene and attribute information of each broadcast template, wherein the broadcast template includes at least one speech combination, and according to the attribute information of each broadcast template and the voice package, the at least one used to select a broadcast template composed of one said scene from one broadcast template,
Broadcast voice generator. 12. The method of claim 11,
The voice generating module,
a text generation sub-module configured to generate broadcast text by filling the broadcast template using the speech;
A voice synthesis sub-module configured to obtain the broadcast voice by performing voice synthesis on the broadcast text by using the tone information in the voice package;
Broadcast voice generator. 10. The method of claim 9,
Further comprising a first excavation module,
The first excavation module includes:
Splicing using preset style keywords and keywords in the scene to obtain search keywords,
selecting a style speech candidate item from the search result text corresponding to the search keyword;
obtaining a style speech by obtaining a result of modifying the style speech candidate item;
to obtain the style speech in the voice package by pre-excavating using the method,
Broadcast voice generator. 10. The method of claim 9,
Further comprising a second excavation module,
The second excavation module includes:
acquiring a knowledge graph related to the voice package,
obtaining a knowledge node matching the scene from the knowledge graph;
generating a knowledge narration corresponding to the scene by using the acquired knowledge node and the narration template corresponding to the scene;
to acquire the knowledge narration in the voice package by excavating in advance using a method,
Broadcast voice generator. In an electronic device,
at least one processor; and
a memory communicatively coupled to the at least one processor;
An instruction executable by the at least one processor is stored in the memory, the instruction is executed by the at least one processor, and the method according to any one of claims 1 to 7 by the at least one processor to do this,
Electronics. In a non-transitory computer-readable recording medium storing computer instructions,
wherein the computer instructions cause the computer to perform the method of any one of claims 1-7.
A non-transitory computer-readable recording medium having computer instructions stored thereon.

Images